Universal Grammar (UG) can be interpreted as a constraint on the form of possible grammars (hypothesis space) or as a constraint on acquisition strategies (selection procedures). In this response to Herschensohn we reiterate the position outlined in Epstein et al. (1996a, r), that in the evaluation of L2 acquisition as a UG-constrained process the former (possible grammars/ knowledge states) is critical, not the latter. Selection procedures, on the other hand, are important in that they may have a bearing on development in language acquisition. We raise the possibility that differences in first and second language acquisition pertaining toboth attainment of the end-state and course of development may derive from differences in selection procedures. We further suggest that for these reasons age effects in the attainment of nativelike proficiency must necessarily be separated from UG effects.